Healthcare

DESCRIPTION

Problem Statement

Dataset Description

The datasets consists of several medical predictor variables and one target variable (Outcome). Predictor variables includes the number of pregnancies the patient has had, their BMI, insulin level, age, and more.

VariablesDescription
PregnanciesNumber of times pregnant
GlucosePlasma glucose concentration in an oral glucose tolerance test
BloodPressureDiastolic blood pressure (mm Hg)
SkinThicknessTriceps skinfold thickness (mm)
InsulinTwo hour serum insulin
BMIBody Mass Index
DiabetesPedigreeFunctionDiabetes pedigree function
AgeAge in years
OutcomeClass variable (either 0 or 1). 268 of 768 values are 1, and the others are 0

Project Task: Week 1

Data Exploration:

  1. Perform descriptive analysis. Understand the variables and their corresponding values. On the columns below, a value of zero does not make sense and thus indicates missing value:
    • Glucose
    • BloodPressure
    • SkinThickness
    • Insulin
    • BMI
  2. Visually explore these variables using histograms. Treat the missing values accordingly.
  3. There are integer and float data type variables in this dataset. Create a count (frequency) plot describing the data types and the count of variables.

Observation : In the above diagrams it looks like for Skin Thickness and Insulin some of the measurements which are displayed as '0' are not recorded only and imputing them either with mean, median, mode will not give us accurate results. So imputing all the missing values (0) in above case with KNN Imputer.

Data Exploration:

  1. Check the balance of the data by plotting the count of outcomes by their value. Describe your findings and plan future course of action.
  2. Create scatter charts between the pair of variables to understand the relationships. Describe your findings.
  3. Perform correlation analysis. Visually explore it using a heat map.

Observation : Pregnancies does not seems to be good predictor of Outcome.
Outcome is Imbalanced class

Observation : There doesn't seems to be good co-relation between any of the variables.

Project Task: Week 2

Data Modeling:

  1. Devise strategies for model building. It is important to decide the right validation framework. Express your thought process.
  2. Apply an appropriate classification algorithm to build a model.
  3. Compare various models with the results from KNN algorithm.
  4. Create a classification report by analyzing sensitivity, specificity, AUC (ROC curve), etc.
Please be descriptive to explain what values of these parameter you have used.

Strategy used : Since the outcome variable is categorical having only 2 values we will try classification models.
The sample data has only 768 rows which needs, which we will split in train and test parts. Since we will require more data to train the models we will go with 20% for the test data.
We will also scale the data with Standard scaler to scale all the values.

Conclusion : Random Forest seems to be a best model among the above.

Data Reporting:

  1. Create a dashboard in tableau by choosing appropriate chart types and metrics useful for the business. The dashboard must entail the following:
    • Pie chart to describe the diabetic or non-diabetic population
    • Scatter charts between relevant variables to analyze the relationships
    • Histogram or frequency charts to analyze the distribution of the data
    • Heatmap of correlation analysis among the relevant variables
    • Create bins of these age values: 20-25, 25-30, 30-35, etc. Analyze different variables for these age brackets using a bubble chart.

Pie%20Chart.png

Scatter-1.png

Scatter-2.png

Hist-Age.png

Hist-BP.png

Hist-BMI.png

Hist-DPF.png

Hist-Glucose.png

Hist-Insulin.png

Hist-skin.png

HeatMap.png

Bubble.png